Data Analytics in Acute Kidney Injury Prediction: Opportunities and Challenges

Abstract

Acute Kidney Injury (AKI) is a common medical condition with a high mortality rate. The incidence of AKI is exceptionally high in hospitalized patients, particularly those suffering from acute illness or postoperative patients. As AKI impacts both patient and financial outcomes, there has been a keen interest the disease. In recent years, AKI and big data synergies have been explored, particularly through electronic health records (EHR), ideal for AKI risk prediction. Due to the massive amount of data in EHR, machine learning (ML) models for data analytics are slowly rising. The application of ML is a promising approach due to its ability to collect EHR data and make predictions on AKI onset accordingly, instead of relying on independent health records. This systematic review aims to identify the opportunities and challenges that arise in integrating data analytics in AKI prediction.